49 lines
1.9 KiB
Python
49 lines
1.9 KiB
Python
import paddleocr
|
|
import os
|
|
|
|
def test_paddleocr_on_pdf(pdf_path):
|
|
print(f"Testing PaddleOCR on: {pdf_path}")
|
|
print(f"File exists: {os.path.exists(pdf_path)}")
|
|
print(f"File size: {os.path.getsize(pdf_path)} bytes")
|
|
|
|
# Initialize PaddleOCR
|
|
ocr = paddleocr.PaddleOCR(use_angle_cls=True, lang='en')
|
|
|
|
try:
|
|
# Run OCR on the PDF
|
|
result = ocr.ocr(pdf_path)
|
|
|
|
print(f"OCR result type: {type(result)}")
|
|
print(f"OCR result length: {len(result) if result else 0}")
|
|
|
|
# Extract and display text
|
|
extracted_text = ""
|
|
if result:
|
|
for page_num, page_result in enumerate(result):
|
|
print(f"\n--- Page {page_num + 1} ---")
|
|
if page_result:
|
|
for line_num, line in enumerate(page_result):
|
|
if line:
|
|
# Each line is [coordinates, (text, confidence)]
|
|
if len(line) >= 2:
|
|
text_info = line[1]
|
|
if len(text_info) >= 1:
|
|
text = text_info[0]
|
|
confidence = text_info[1] if len(text_info) > 1 else "N/A"
|
|
extracted_text += text + " "
|
|
print(f"Line {line_num}: '{text}' (confidence: {confidence})")
|
|
else:
|
|
print("No text detected on this page")
|
|
|
|
print(f"\nTotal extracted text length: {len(extracted_text)}")
|
|
print(f"Extracted text: '{extracted_text.strip()}'")
|
|
|
|
return extracted_text.strip()
|
|
|
|
except Exception as e:
|
|
print(f"Error during OCR: {e}")
|
|
return ""
|
|
|
|
if __name__ == "__main__":
|
|
extracted = test_paddleocr_on_pdf('ocr.pdf')
|
|
print(f"\nFinal result: Text extracted: {len(extracted) > 0}") |